EP2360630A1 - Quality assurance of industrial production and assembly processes - Google Patents

Abstract

The method involves systematically detecting and analyzing product data of the production of downstream production life cycle sections. The analyzed product data are directly used for the quality assurance of the industrial production or assembly process.

Description

Quality assurance of industrial production and assembly processes

The invention relates to a method for quality assurance of an industrial production process or an assembly process for the production of products, as well as a device for carrying out the method.

The characterization of product and quality characteristics in industrial processes (e.g., production, assembly) has limitations in terms of economy and practicability. This means in practice e.g. that in processes (depending on the application and after estimation of admissibility) not all technically feasible measurements are carried out, only randomly measured or processes are driven primarily due to more or less comprehensive predominantly plant-local experience.

Mastering the problem occurs e.g. by prior estimation of the required measuring effort (measured variables, measuring instruments, measuring intensity, etc.) and, if necessary, by adaptation or changes of the industrial process in case of complaints or complaints from customers.

Although product complaints and complaints (eg complaints, feedback, feedback) from customers are often queried and analyzed by companies as part of CRM (Customer Relationship Management), changes to parameters in the underlying manufacturing process of the product can only be delayed Consideration.

The object of the present invention is to provide an improved method for the quality assurance of industrial manufacturing processes based on feedback data from consumers.

The object is achieved by a method for quality assurance of an industrial production process or an assembly process for the production of products, wherein product data of the production of downstream product life cycle sections are systematically recorded and analyzed and used directly for the quality assurance of the industrial production or assembly process. This allows the structured use of "indirect" information to improve the experience base in the industrial production or assembly process. For this purpose, all relevant available information from downstream life cycle sections (eg processing, use, disposal / recycling) is collected and used for ongoing quality control. The downstream data (ie the information from downstream life cycle sections) accumulates during normal processing. The technical realization of the data feedback (feedback) can be carried out, for example, preferably via online IT interfaces with which data correspondingly defined between manufacturer and processor (consumer) are automatically transferred between the processing systems of the processor and a quality management system of the manufacturer. A collaborative CRM system, which for example links the quality management systems of suppliers and customers in terms of data technology, can also enable systematic and direct data feedback. The invention allows the utilization of "indirect", ie process-remote information for the quality control of a production. This is especially in the environment of cost-sensitive bulk goods or complex measurements of immediate economic value, contributes to a stabilization of product quality and uses synergetic (often) existing data anyway.

A first advantageous embodiment of the present invention consists of the fact that for each product and for each half product used for the production there is a data record in which the data available from later life cycle sections are stored. This enables a systematic and unambiguous assignment of feedback data of downstream product life cycle sections.

A further advantageous embodiment of the present invention consists in that the data records for a product are collected and aggregated into a data record for the corresponding product. Data collection for a product allows early detection of a problem. If e.g. If processor A (for example, a customer following the product in the value chain) encounters a problem, processors B and C can be informed early. But if the same error is reported by several processors, this is an indication that production must be readjusted immediately.

A further advantageous embodiment of the present invention is that the data records for a product are aggregated into a data record for a corresponding product type and / or product model and / or a variety. The data collection for a product type or variety allows for a long-term optimization of the type or even, if necessary, the achievement of a status of "farm trial" for the variety. The production parameters for a long-term proven type or variety can then be reused with a "clear conscience" for a follow type or a following sort or serve as a basis.

A further advantageous embodiment of the present invention is that the respective data sets for semi-products, products, product types and product models are each examined for changes and the differences are related to the respective production or assembly conditions. By correlating the datasets (the datasets also represent the quality of the respective semi-finished products, products, product types or product models) with the particular production or assembly conditions of the manufacture, an objective relationship and also a conclusion between the production and assembly conditions can be made and the quality of the products produced and proven.

A further advantageous embodiment of the present invention is that the data from the downstream life cycle sections originate from the quality and asset management of downstream internal company processes and / or from sales information. Sales information typically involves direct customer or market feedback about a product, and thus represents information that can be dedicated to the particular product manufacturing process. Further advantages arise when the quality and asset management of company-internal processes are data-linked or integrated into a PLM system (Product Lifecycle Management). A PLM system is an information technology (IT) system that unites all the data involved in the development, production, warehousing and distribution of a product stored, managed and retrieved. Advantageously, all areas or systems which come into contact with a product access a common database. Planning (ERP / PPS), construction (CAD), calculation (CAE) and production (CAM), controlling, sales, service and quality assurance are integrated into the data. Feedback from one area can thus be considered immediately in upstream or downstream areas.

A further advantageous embodiment of the present invention is that the data from the downstream life cycle sections originate from the quality and asset management of downstream external processes and / or from customer relationship tools and / or independent experts' test reports. This allows for systematic product-related data collection from third parties who use the product and thus enables rapid, targeted response to external feedback. Data mining tools and methods (e.g., Internet crawlers or search engines) can also be used to obtain product-specific information for use in improving the manufacturing process.

A further advantageous embodiment of the present invention consists in that the method for quality assurance in coil production (bundles) is used, the quality management system of the coil manufacturer receiving from the quality management system of a downstream processor automatically defined quality data of products of the processor via an online interface , A bundle-dependent changed reject rate, a change in cutter life / press forces or a changed chip formation during turning (longer, shorter, more chapped, etc.) are valuable hints a change in coil properties that can be used to optimize coil production. For example, the technical implementation may preferably take the form of online IT interfaces, with which data defined accordingly between coil manufacturer and processor are automatically transferred between the processing systems of the processors and a quality management system of the steel producer.

A further advantageous embodiment of the present invention is that the quality data of the downstream processor (e.g., car producer) is measurement data and / or sensor data taken directly into the quality management system of the coil manufacturer (e.g., steel mill). Measurement data and sensor data from a downstream processor are data that is generated prior to a rollout (for example, delivery to customers) of the product. If product problems are systematically detected before the product is shipped and returned to the manufacturer, the cost of resolving the problem is lower than after a rollout.

A further advantageous embodiment of the present invention is that the product is provided with an RDID tag and the recycling properties are compared at the end of life of the product with the data of the semi-products used to produce the product. Objects marked with RFID tags permit an assignment until the object melts again and, in principle, make it possible to compare the recycling properties at the end of the life cycle of a product (eg passenger car, beverage can, etc.) with the data of the coil used for its production. A further advantageous embodiment of the present invention is that the data of downstream external processors and / or users of the product are provided directly online to the operator of the upstream product manufacturing process. Advantageously, the external processor data is provided directly in the product manufacturing process (ie, at the "shop floor" level). For this purpose, it is necessary that defined (coordinated, standardized or standardized) data interfaces exist between the operator of the production process and the downstream processors / users. In steelmaking, for example, an interface could have the following form: (name, unit, detailed description), with the following exemplary data contents: 'RollForceFm1', 'N', 'technological roll force in state 1 of Finishing Mill'. Advantageously, the data connection between the operator of the manufacturing process and the downstream processors / users via encryption mechanisms (eg PGP) is secured.

A further advantageous embodiment of the present invention is that the data of a downstream external processor and / or user are provided via a neutral third party to the operator of the upstream product manufacturing process. By storing and forwarding the data by a 'neutral' third party, e.g. ensure anonymization of the data.

The object is further achieved by a device for carrying out one of the above methods. Advantageously, the arrangement is based on a data technology integration of the manufacturing process with downstream in-house and / or external processes of the product life cycle. The device can eg as a PLM system (Product Lifecycle Management / product life cycle management). A PLM system is an information technology (IT) system that can uniformly store, manage and retrieve all data related to the development, production, warehousing and distribution of a product. Advantageously, all areas or systems related to a product access a common database. Planning (ERP / PPS), design (CAD), calculation (CAE), production (CAM), controlling, sales, service and quality assurance are integrated into the data. External systems (eg from customers or suppliers) can also be integrated. Integration can take place via LAN or WLAN technology, but also via the Internet or Intranet.

An embodiment of the invention is illustrated in the drawing and will be explained below.

FIG 1

ein beispielhaftes und schematisches Übersichtsbild zu Eigenschaftsräumen von Stahlprodukten,

FIG 2

ein schematisches Übersichtsbild zur Qualitätssteuerung bei der Erzeugung von Stahlprodukten,

FIG 3

ein schematisches Übersichtsbild zur Qualitätssteuerung unter Nutzung nachgelagerter Lebenszyklusabschnitte,

FIG 4

ein schematisches Übersichtsbild zur Aggregation von der Produktion nachgelagerten Daten, und

FIG 5

ein beispielhaftes schematisches Übersichtsbild zur Dokumentation der Produktionsergebnisse.

Showing:

FIG. 1

an exemplary and schematic overview of the properties of steel products,

FIG. 2

a schematic overview image for quality control in the production of steel products,

FIG. 3

a schematic overview image for quality control using downstream life cycle sections,

FIG. 4

a schematic overview image for the aggregation of downstream production data, and

FIG. 5

an exemplary schematic overview picture for the documentation of the production results.

FIG. 1 shows an exemplary and schematic overview image to property spaces of steel products. The characterization of product properties in industrial processes encounters limits in terms of cost-effectiveness and practicability. This means in practice, for example, that in processes (depending on the application and after estimation of admissibility) not all technically possible measurements are performed, only randomly measured (at measuring points M1 to M7) or processes primarily due to more or less comprehensive predominantly system-local experience values are driven.

A concrete application example is the production of flat steel products, eg coils P1 to Pn, in which an exact determination of the rolling textures (anisotropic distribution of the crystallites depending on the rolling process) is technically possible, but because of the associated effort (sampling, metallographic processing, radiographic Examination, data evaluation and interpretation) only in problem or special cases (eg electrical sheets) takes place. This results in particular for "uncritical" and / or low-margin products, the problem that the quality of the products P1 to Pn, although with a certain probability in the confidence interval, a leaving the permissible range is not systematically recognized. In particular, this problem also concerns transitional states (production start-up, drafting operation, etc.) or product (ion) conversions. The evaluation of the measuring points M1 to M7 also gives an indication of the quality of the variety S. belonging to the products P1 to Pn.

FIG. 2 shows a schematic overview image for quality control in the production of steel products, which provides a first approach to the possible control of in FIG. 1 indicated measuring problem. The problem is mastered by prior estimation of the required measuring effort (measured quantities, devices, intensity, etc.) and, if necessary, adjustment in case of complaints. However, the activities take place only in the immediate (technical) environment of production. In FIG. 2 the process SE1 for the production of steel products (eg coils) is regulated and monitored by a quality control QS1. The quality control QS1 evaluates sensor and measurement data from the steelmaking process SE1 and acts on the process SE1 by changing or adapting process parameters or process variables. The representation after FIG. 2 But it provides a control loop that only works within the actual production process.

FIG. 3 shows a schematic overview image for quality control of a manufacturing process (eg production or assembly process) using downstream life cycle sections of the manufacturing process. The invention is based on the structured use of "immediate" information to improve the experience base of the manufacturing process of a product. For this purpose, all relevant available information from downstream life cycle sections (eg processing WV1, use N, disposal / recycling R) are collected and used for the ongoing quality control QS2 of the production process SE2.

Data acquisition and processing

In the other sections of the lifecycle, a large amount of data accumulates in different systems. In milling machines, for example, the service life of the tools are recorded and recorded this basis controlled their exchange. For pressing, forming forces are measured. When machining, the shape of the chip is an essential factor for adjusting the machines. When welding, the appearance and continuity of the molten areas is an important factor. In a pickling plant, the reaction behavior of the surface is a directly material-dependent quantity. Further data is obtained when using products, such as TÜV reports on the individual car models.

This makes it possible to use implicit knowledge from later process steps or lifecycle sections for evaluating the product, without costly measurements that are explicitly tailored to an effect, preferably for products with short life cycles, which still allow subsequent control of production SE2.

• Qualitäts- und Assetmanagement nachgelagerter firmen-interner Prozesse (z.B. Verzinkungsanlage, Stanzerei, Lackiererei),
• Qualitäts- und Assetmanagement nachgelagerter externer Prozesse (z.B. kooperierender Kunde, Recyclingfirma),
• Customer Relationship-Werkzeugen (z.B. Reklamationen, Supportanfragen, Feedbackbögen),
• Vertriebsinformationen (z.B. Änderungen / Auffälligkeiten im Bestellverhalten).

For example, data from downstream life cycle sections can be fed from:

• Quality and asset management of downstream internal company processes (eg galvanizing plant, stamping shop, paint shop),
• Quality and asset management of downstream external processes (eg cooperating customer, recycling company),
• Customer Relationship tools (eg complaints, support requests, feedback forms),
• Sales information (eg changes / abnormalities in order behavior).

• Messergebnisse im Rahmen von Wareneingangsprüfungen,
• Auffälligkeiten beim Tiefziehverhalten (z.B. Fehler als Hinweis zu Walztextur),
• Erhöhter Werkzeugverschleiss (z.B. Fräskopfstandzeit / - schäden als Hinweis zu Änderungen in der Härte),
• Lackierfehler (z.B. Hinweis zu Benetzungseigenschaften / Verschmutzung),
• Korrosion (z.B. Hinweis zu chemischer Zusammensetzung, Mikrorissbildung durch Cu-Anteil u.ä.),
• Oberflächenfehler,
• Ausschußquoten,
• Verarbeitbarkeit (z.B. Art der Spanbildung bei spanender Bearbeitung).

Interested data can be eg:

• Measurement results in the context of incoming goods inspections,
• Abnormalities in thermoforming behavior (eg error as an indication of rolling texture),
• Increased tool wear (eg milling head life / damage as an indication of changes in hardness),
• Painting defects (eg information on wetting properties / contamination),
• Corrosion (eg information on chemical composition, microcracking due to Cu content, etc.),
• Surface defects,
• Committee quotas,
• Workability (eg type of chip formation during machining).

In particular, information that results from the increasing possibilities of tracking through RFID tags is also interesting. In principle, after the correspondingly marked objects permit an association until remelting, it is possible to compare the recycling properties at the end of life of a product (e.g., car, beverage can, etc.) with the data of the coil used to make it.

FIG. 4 shows a schematic overview image for the aggregation of downstream production data. An essential feature of the invention is that it is primarily about the use of downstream data, resulting in the context of a normal processing, etc. A bundle-dependent changed reject rate, a change in cutter life / press forces or a changed chip formation during turning (longer, shorter, more chapped, etc.) are valuable indications for a change of coil properties, which can be used to optimize the coil production ,

The technical realization can e.g. preferably take the form of on-line IT interfaces, with which according to defined data between coil manufacturer and processor automatically transferred between the processing systems of the processors and a quality management system of the steel producer. Another variant is the use of aggregated information, e.g. the o.g. TÜV reports or evaluations that are carried out by recycling companies and must be entered manually.

The data thus obtained can now be e.g. By comparing different bundles in the same application, they are examined for changes and related to differences in production conditions.

In the exemplary flowchart according to FIG. 4 the production downstream data VRD are systematically recorded, filtered according to relevance to a particular manufacturer (eg coil manufacturer) (F) and provided in a defined data format (eg XML) to the manufacturer as transfer data UD. The transfer data UD are adapted to the context of the manufacturer (AN) and from this a supplementary product description ECD (eg coil description) is created. The adaptation to the context of coil production can, for example, consist in the derivation of measured values and key figures. The next step is the aggregation AG of the product data to create a complementary product or variety description ESD. Coil production aggregates data from multiple coils of the same product.

A further advantageous embodiment of the invention consists in the extension to eg the possibility for correlation formation with the production data, eg as part of a quality management or "data mining" tool, as well as the integration into existing warning and reporting systems and making available in the immediate production environment (eg through a "processing data view"). In particular, this type of collection also allows the use of insights from a (possibly more sophisticated) application for other applications ("..that worked safely with x, so it will work well with y").

FIG. 5 shows an exemplary schematic overview image for the documentation of the production results. The documentation is based on data from the manufacturing process SE3 and on downstream internal and / or external data NGD (shown by dotted arrow). Downstream external data may be eg consumer information (eg from trade journals or test reports) or TÜV reports. The product documentation is divided into a product documentation PDE per unit produced and a product documentation PDS for the associated grade or product type.

Product documentation PDE (per unit produced):

• For each product unit (e.g., coil), a record is kept of available data from later lifecycle stages (storage of data from data collection and rendering).

Product documentation PDS (per product / product type / variety):

• For each product (for example, variety A), a data record for external data is maintained in which the contents of the product documentation PDE are aggregated per unit produced. The aggregation can be done, for example, in the following way:
  • Collection of use cases / processing modes for which the material is used
  • Filling of property spaces / development of tolerance bands, eg: the material was machined with milling cutters with cutting speeds from u to vm / s and feed rates from x to y mm / s.

• Reduzierung von Eigenschaften, die von Anwendern nicht genutzt werden,
• Erfahrungsbasierte Einsetzbarkeit von Minder-Qualitäten für spezielle Anwendungen (Einzelfall-Qualifizierung)
• Optimierung von Produkten für Haupt-Einsatzbereiche.

This approach results in a product-specific collection of information that complements, confirms and helps validate the assumptions about the interpretation of the measurements. Furthermore, this application-specific collection of information makes it possible to optimize products and offers, for example by:

• Reduction of properties that are not used by users
• Experience-based applicability of Minder qualities for special applications (case-by-case qualification)
• Optimization of products for main applications.

A method and apparatus for quality assurance of an industrial production process or assembly process for manufacturing products, wherein product data of the production of downstream product life cycle portions are systematically collected and analyzed and used directly for quality assurance of the industrial production or assembly process. In particular, the device and the method can be integrated into a PLM system (Product Lifecycle Management) or implemented as a PLM system.

reference numeral

M1 - M7

measuring point

S

variety

P1 - Pn

product

QS1, QS2

quality control

SE1 - SE3

steelmaking

WV1, WV2

processing

N

use

R

recycling

F

filtering

AT

Adaptation

AG

aggregation

VRD

Processing relevant data

UD

Transfer Data

ECD

Additional coil description

ESD

Supplementary variety description

NGD

Downstream data

PDE

Product documentation unit

PDS

Product Documentation Variety

Claims (14)

Quality Assurance Procedures (QS1, QS2) of an industrial production process (SE1 - SE3) or an assembly process for the production of products (P1 - Pn), whereby product data for the production of downstream product life cycle segments (WV1, WV2, N, R) are systematically collected and analyzed; for quality assurance (QS1, QS2) of the industrial production or assembly process (SE1 - SE3). Method according to claim 1,
characterized in that
for each product (P1-Pn) and for each half-product used for production, a data record exists in which the data available from later lifecycle sections (WV1, WV2, N, R) are stored. Method according to claim 1 or 2,
characterized in that the records for a product (P1 - Pn) are aggregated to form a record for the corresponding product. Method according to claim 1 or 2,
characterized in that the data records for a product (P1 - Pn) are aggregated to a data record for a corresponding product type and / or product model and / or a variety (S). Method according to one of the preceding claims, characterized in that the corresponding data sets for semi-products, products (P1-Pn), product types and product models are each subject to changes be examined and the differences related to the respective production or assembly conditions. Method according to one of the preceding claims, characterized in that the data from the downstream life cycle sections (WV1, WV2, N, R) originate from the quality and asset management of downstream internal company processes and / or from sales information. Method according to one of the preceding claims, characterized in that the data from the downstream life cycle sections (WV1, WV2, N, R) from the quality and asset management of downstream company-external processes and / or customer relationship tools and / or Audit reports by independent experts. Method according to one of the preceding claims, characterized in that the method for quality assurance in the coil production (coils) is used, wherein the quality management system (QS1, QS2) of the coil manufacturer from the quality management system of a downstream processor (WV1, WV2, N, R) receives automatically defined quality data of products of the processor (WV1, WV2, N, R) via an online interface. Method according to claim 8,
characterized in that the quality data of the downstream processor are measurement data and / or sensor data which are stored directly in the Quality management system (QS1, QS2) of the coil manufacturer. Method according to one of the preceding claims, wherein the product (P1-Pn) is provided with an RDID tag and the recycling properties at the end of life cycle (R) of the product (P1-Pn) with the data for the production of the product (P1-Pn) used half-products are compared. Method according to one of the preceding claims, wherein the data of downstream external processors and / or users of the product are provided directly online to the operator of the upstream product manufacturing process. The method of claim 11, wherein
the data of a downstream external processor and / or user are provided via a neutral third party to the operator of the upstream product manufacturing process. Apparatus for carrying out a method according to one of claims 1 to 12. Device according to claim 13,
characterized in that
the device is a PLM system.

Images